Does comparative advantage explain legislators' support for trade liberalization? We use data on potential crop yields as determined by weather and soil characteristics to derive a new, plausibly exogenous measure of comparative advantage in agriculture for each district in the US. Evidence shows that comparative advantage in agriculture predicts how legislators vote on the rati cation of preferential trade agreements in Congress. We show that legislators in districts with high agricultural comparative advantage are more likely to mention that trade agreements are good for agriculture in House oor debates preceding roll-call votes on their rati cations. Individuals living in the same districts are also more likely to support free trade. Our analysis and results contribute to the literature on the political economy of trade and its distributional consequences, and to our understanding of the economic determinants of legislators voting decisions.